Cukrzyca typu 2
Rokowania, prognozy i postęp choroby

Cukrzyca typu 2, będąca przewlekłym zaburzeniem metabolicznym charakteryzującym się hiperglikemią wynikającą z insulinooporności i upośledzonej sekrecji insuliny, stanowi globalne wyzwanie zdrowotne z prognozowaną liczbą chorych sięgającą 350 milionów do 2030 roku. Wczesne wykrycie osób z wysokim ryzykiem rozwoju choroby jest kluczowe dla skutecznej profilaktyki i leczenia. Modele predykcyjne, uwzględniające tradycyjne czynniki ryzyka takie jak wiek, płeć, BMI, poziom HbA1c, a także nowoczesne biomarkery zapalenia (IL-10), stresu oksydacyjnego (8-izoprostan, GSSG) i dysfunkcji mitochondrialnej (humanina, P66Shc), poprawiają identyfikację pacjentów zagrożonych rozwojem cukrzycy. Jednakże większość modeli wykazuje umiarkowaną zdolność dyskryminacyjną (np. AUC 0,73–0,79) i wymaga dalszej kalibracji oraz walidacji zewnętrznej, aby zwiększyć ich wiarygodność i użyteczność kliniczną.

Prognozy dla cukrzycy typu 2

Cukrzyca typu 2 jest przewlekłym zaburzeniem metabolicznym charakteryzującym się nieprawidłowo podwyższonym poziomem glukozy we krwi, wynikającym z insulinooporności i zmniejszonej produkcji insuliny przez trzustkę. Stanowi poważne wyzwanie dla systemów opieki zdrowotnej na całym świecie, a Światowa Organizacja Zdrowia szacuje, że do 2030 roku liczba osób z tą chorobą osiągnie około 350 milionów.1 Ponad 90% osób chorujących na cukrzycę cierpi na typ 2, którego rozwój jest uwarunkowany czynnikami społeczno-ekonomicznymi, demograficznymi, środowiskowymi i genetycznymi.2 Wczesne rozpoznanie pacjentów z niezdiagnozowaną cukrzycą typu 2 lub osób ze zwiększonym ryzykiem rozwoju tej choroby stanowi istotne wyzwanie dla współczesnej medycyny.3

Modele predykcyjne w ocenie ryzyka

Modele predykcyjne ryzyka mają znaczny potencjał w procesie podejmowania decyzji dotyczących postępowania klinicznego z pacjentem. Interwencje medyczne lub zmiany stylu życia mogą być skuteczniej ukierunkowane na osoby ze zwiększonym ryzykiem rozwoju choroby.4 Wykorzystanie modeli prognostycznych może pomóc w identyfikacji osób, u których rozwinie się cukrzyca typu 2, co pozwala na bardziej efektywne ukierunkowanie działań profilaktycznych.5

W ostatnich latach opracowano wiele modeli predykcyjnych do oceny ryzyka rozwoju cukrzycy typu 2. Większość podstawowych modeli predykcyjnych może identyfikować osoby o wysokim ryzyku rozwoju cukrzycy w okresie od pięciu do dziesięciu lat. Modele uwzględniające biomarkery klasyfikują przypadki nieco lepiej niż modele podstawowe, chociaż większość modeli przeszacowuje rzeczywiste ryzyko cukrzycy.6

Badania wykazały, że modele predykcyjne dobrze sprawdzają się w identyfikacji osób z wysokim ryzykiem przyszłej cukrzycy, co jest pierwszym warunkiem stosowania takich modeli w praktyce zgodnie z obecnymi zaleceniami.7 Jednakże znaczące odchylenia między przewidywanym a obserwowanym ryzykiem pozostają dla większości modeli.8

Czynniki prognostyczne w cukrzycy typu 2

W analizie czynników prognostycznych dla cukrzycy typu 2 zidentyfikowano szereg istotnych parametrów. Oprócz dobrze znanych czynników predykcyjnych stanu przedcukrzycowego, takich jak wiek, płeć i BMI, zaobserwowano, że miary samooceny stylu życia, zdrowia i wsparcia społecznego są ważnymi i modyfikowalnymi predyktorami cukrzycy.910

W badaniach nad pacjentami ze stanem przedcukrzycowym ustalono, że ostateczny model predykcyjny powinien uwzględniać poziom HbA1c, wiek, płeć, wskaźnik masy ciała (BMI), stosowanie leków przeciwnadciśnieniowych, choroby trzustki, nowotwory, samodzielnie zgłaszaną dietę, zalecenia lekarza dotyczące utraty masy ciała lub zmiany nawyków żywieniowych, posiadanie osób do rozmowy oraz samoocenę stanu zdrowia.11

Badania z zastosowaniem zaawansowanych technik uczenia maszynowego zidentyfikowały nową kombinację biomarkerów, w tym interleukiny-10 (IL-10), 8-izoprostanu, humaniny (HN) i utlenionego glutationu (GSSG), które okazały się bardziej wpływowe niż tradycyjne biomarkery w prognozowaniu cukrzycy typu 2.12 Pięcioma najważniejszymi predyktorami pod względem wartości DIFFI były IL-10, 8-izoprostan, GSSG, HN i P66Shc, podczas gdy najniższe wyniki uzyskały poziom glukozy we krwi i trójglicerydy.13

W kontekście czynników predykcyjnych szczególnie istotne wydają się: poziom glukozy, wskaźnik masy ciała, funkcja rodowodu cukrzycowego i wiek, które są konsekwentnie identyfikowane jako najlepsze i najczęściej dokładne predyktory wyników.14

Ryzyko progresji ze stanu przedcukrzycowego

Stan przedcukrzycowy zwiększa ryzyko rozwoju cukrzycy typu 2. Badania wskazują, że jedna na pięć osób ze stanu przedcukrzycowego będzie rozwijać cukrzycę typu 2 zdefiniowaną wartością HbA1c w ciągu 5 lat od pierwszej diagnozy przedcukrzycowej.1516 W trakcie mediany obserwacji wynoszącej 2,7 roku, 11,8% osób ze stanem przedcukrzycowym postępowało do cukrzycy typu 2, a 10,1% zmarło.17

Pomimo identyfikacji osób ze stanem przedcukrzycowym o wysokim ryzyku, czasowa krzywa ROC (Area Under the Curve) wynosiła tylko 73 (95% CI 71 do 74) dla cukrzycy zdefiniowanej wartością HbA1c i 79 (95% CI 78 do 81) dla rozpoczęcia leczenia obniżającego poziom glukozy.1819 Wskazuje to na umiarkowaną zdolność dyskryminacyjną modeli predykcyjnych.

Powikłania i rokowanie długoterminowe

Cukrzyca typu 2 jest chorobą przewlekłą, co oznacza, że wymaga zarządzania przez całe życie pacjenta.20 Rokowanie dla pacjentów z cukrzycą typu 2 zależy od wielu czynników, a nieleczona lub niedostatecznie kontrolowana choroba może prowadzić do szeregu stanów zdrowotnych.21

Powikłania sercowo-naczyniowe

Powikłania sercowo-naczyniowe stanowią jedno z najpoważniejszych zagrożeń dla pacjentów z cukrzycą typu 2. Badania porównujące wydajność 22 skal oceny ryzyka sercowo-naczyniowego u pacjentów z cukrzycą typu 2 wykazały, że skala RECODE, opracowana dla osób z cukrzycą typu 2, wykazała najlepszą wydajność zarówno dla chorób układu krążenia (współczynnik c-statystyki 0,731 (0,728; 0,734)), jak i poszerzonych zdarzeń sercowo-naczyniowych (0,732 (0,729; 0,735)).22

Interesujące jest, że ani populacja, w której opracowano skalę, ani pierwotnie przewidywany wynik nie wpływały na wydajność skali. Nachylenia kalibracyjne (gdzie 1 oznacza doskonałą kalibrację) wahały się od 0,38 (95% CI 0,37; 0,39) do 1,05 (95% CI 1,03; 1,07).23 Prosty, specyficzny dla populacji proces rekalibracji znacznie poprawił wydajność, wahając się między 0,98 a 1,03.24

Warto zauważyć, że skale ryzyka działały słabo u osób z istniejącymi chorobami układu krążenia (współczynnik c-statystyki 0,55), a skale z większą liczbą predyktorów nie działały lepiej: dla rozszerzonych zdarzeń sercowo-naczyniowych QRISK3 (19 zmiennych) współczynnik c-statystyki wynosił 0,69 (95% CI 0,68; 0,69), w porównaniu do CHD Basic (8 zmiennych) 0,71 (95% CI 0,70; 0,71).25

Wskaźniki jakości leczenia a wyniki sercowo-naczyniowe

Badania wykazały, że wskaźniki jakości mierzące aktualny status leczenia obniżającego poziom lipidów i albuminurię przewidywały niższe ryzyko poważnych zdarzeń sercowo-naczyniowych i śmiertelności u pacjentów z cukrzycą w praktyce ogólnej.26 Wskaźniki jakości dla leczenia obniżającego poziom glukozy powinny być stosowane tylko dla ograniczonych populacji z podwyższonym poziomem HbA1c.27

Co ciekawe, testowane wskaźniki leczenia obniżającego ciśnienie krwi nie przewidywały wyników pacjenta.28 Żaden ze wskaźników jakości mierzących leczenie obniżające ciśnienie krwi lub intensyfikację leczenia nie był predyktorem twardych punktów końcowych.29

Wskaźniki jakości mierzące status leczenia obniżającego poziom lipidów i albuminurię można rozważyć do wdrożenia do zestawów wskaźników jakości, ponieważ wskaźniki te wydają się skutkować mniejszą liczbą zdarzeń sercowo-naczyniowych.30 Wskaźnik mierzący status leczenia obniżającego poziom glukozy powinien być ograniczony tylko do pacjentów z podwyższonym HbA1c.31

Nowoczesne metody predykcji wyników

W ostatnich latach obserwuje się rosnące zainteresowanie wykorzystaniem zaawansowanych technik uczenia maszynowego i głębokich sieci neuronowych do prognozowania wyników cukrzycy typu 2. Badania porównawcze wykazały, że głębokie sztuczne sieci neuronowe (ANN) przewyższają inne testowane klasyfikatory uczenia maszynowego, osiągając dokładność 95,14%.3233

Jednak inne badania nie wykazały klinicznie istotnej poprawy przy zastosowaniu bardziej wyrafinowanych modeli predykcyjnych w porównaniu do standardowych technik regresji.34 Większość testów przesiewowych dla cukrzycy typu 2 stosowanych obecnie została opracowana przy użyciu metod regresji wielowymiarowej, które są często dalej upraszczane, aby umożliwić przekształcenie w formułę punktową.35

Badania nad metodami uczenia maszynowego w celu wczesnego przewidywania upośledzonej glikemii na czczo (IFG) i poziomów glukozy w osoczu na czczo (FPGL) nie wykazały klinicznie istotnej poprawy przy zastosowaniu modeli opartych na uczeniu maszynowym w porównaniu do bardziej konwencjonalnych modeli regresji pod względem wydajności predykcyjnej.36

Wśród wszystkich klasyfikatorów, GAMBoost i GAMLOESS zapewniły najlepsze wyniki w analizie. GAMBoost uzyskał lepszą wydajność niż GAMLOESS pod względem dokładności i czułości, podczas gdy GAMLOESS wykazał lepsze wyniki dla AUROC i swoistości.37

Ograniczenia istniejących modeli predykcyjnych

Pomimo dużej liczby opracowywanych modeli predykcyjnych ryzyka, tylko bardzo niewielka mniejszość z nich jest rutynowo wykorzystywana w praktyce klinicznej.38 Rośnie obawa, że większość modeli predykcyjnych ryzyka jest słabo rozwinięta, ponieważ opiera się na małym i nieodpowiednim doborze kohorty, wątpliwym postępowaniu z ciągłymi predyktorami ryzyka, nieodpowiednim traktowaniu brakujących danych, stosowaniu wadliwych lub nieodpowiednich metod statystycznych oraz, ostatecznie, braku przejrzystego raportowania kroków podjętych w celu wyprowadzenia modelu.39

Problemy metodologiczne

Systematyczny przegląd opublikowanych badań wskazuje na liczne niedociągnięcia metodologiczne i ogólnie niski poziom raportowania w badaniach, w których opracowano modele predykcyjne ryzyka do wykrywania występującej lub pojawiającej się cukrzycy typu 2.40

Zaobserwowano powszechne stosowanie złych metod, które mogą zagrozić rozwojowi modelu, w tym jednowymiarowe wstępne przesiewanie zmiennych, kategoryzację ciągłych predyktorów ryzyka i złe postępowanie z brakującymi danymi.41 Stosowanie złych metod wpływa na wiarygodność modelu predykcyjnego i ostatecznie zagraża dokładności szacunków prawdopodobieństwa występowania niezdiagnozowanej cukrzycy typu 2 lub przewidywanego ryzyka rozwoju cukrzycy typu 2.42

Potrzeba walidacji zewnętrznej

Ogólnie rzecz biorąc, pomimo nowych czynników ryzyka lub nowych aspektów metodologicznych, nowo opracowany model nie zwiększył naszej zdolności do badania przesiewowego/przewidywania cukrzycy typu 2, głównie w części analitycznej. Wynikało to z braku zewnętrznej walidacji modeli predykcyjnych.43

Stawia to pytanie, czy możemy polegać na obecnych modelach predykcyjnych, czy powinniśmy rozwijać nowe modele. Innym głównym problemem jest to, że nowo opracowany model może być łatwo zignorowany, jeśli nie ma dodatkowej wartości dla decydentów w zakresie polityki zdrowotnej lub klinicystów.44

Perspektywy dla pacjentów i personelu medycznego

Biorąc pod uwagę szeroki zakres prezentacji i rozwoju chorób współistniejących w cukrzycy typu 2, leczenie i opieka nad pacjentami mogą być znacznie ulepszone, jeśli oznaki prognostyczne zostaną wykorzystane do lepszej sub-kategoryzacji pacjentów z cukrzycą typu 2.45

Potencjał interwencji profilaktycznych

Mimo że cukrzyca typu 2 ma poważny wpływ na zdrowie publiczne, możliwe jest zmniejszenie jej wpływu poprzez podejmowanie środków zapobiegawczych dla cukrzycy typu 2 oraz zapewnienie wczesnej diagnozy i właściwej opieki dla wszystkich typów cukrzycy. Te środki mogą pomóc ludziom żyjącym z tym schorzeniem uniknąć lub opóźnić powikłania.46

Wczesna identyfikacja osób zagrożonych rozwojem cukrzycy typu 2 jest priorytetem dla zapobiegania długoterminowym powikłaniom choroby.47 Włączenie biomarkerów stresu oksydacyjnego, zapalenia i dysfunkcji mitochondrialnej poprawiło wydajność wszystkich metryk w porównaniu do modelowania predykcyjnego tylko z tradycyjnymi biomarkerami poziomu glukozy we krwi, BMI i trójglicerydów. Największy wzrost wydajności zaobserwowano dla wskaźników czułości i F1.48

Personalizacja leczenia

Rozwój zindywidualizowanych podejść do leczenia cukrzycy typu 2 może poprawić wyniki pacjentów. Badania wskazują, że typowe monitorowanie ryzyka cukrzycy typu 2 poprzez poziom glukozy we krwi może nie dostarczać kompleksowego obrazu progresji choroby.49

Wpływowe biomarkery zidentyfikowane w badaniach, takie jak IL-10, 8-izoprostan, GSSG, HN i P66Shc, ujawniają potencjał biomarkerów zapalenia, stresu oksydacyjnego i dysfunkcji mitochondrialnej do służenia jako wskazówka dla ukierunkowanej, spersonalizowanej interwencji w zapobieganiu występowaniu cukrzycy typu 2.50

W kontekście leczenia bariatrycznego, opracowano model predykcyjny dla wyników utraty masy ciała po laparoskopowej rękawowej gastrektomii (LSG) przez integrację trzech kluczowych wskaźników: spoczynkowego wydatku energetycznego na masę ciała (REE/BW), wskaźnika beztłuszczowej masy ciała (FFMI) i obwodu talii (WC). Kombinacja tych wskaźników zapewnia stosunkowo dokładną przedoperacyjną prognozę wyników utraty masy ciała rok po LSG.51

To holistyczne podejście do prognozowania i leczenia cukrzycy typu 2 uwzględniające szerokie spektrum biomarkerów i czynników klinicznych oferuje nadzieję na lepsze wyniki leczenia i jakość życia dla milionów osób dotkniętych tą chorobą na całym świecie.

Kolejne rozdziały

Zapraszamy do dalszego czytania naszego leksykonu.

Wybierz kolejny rozdział z menu poniżej, aby otworzyć nową podstronę kompedium wiedzy i uzyskać szczegółowe informację o leku, substancji lub chorobie.

  1. 10.04.2026
  2. www.leksykon.com.pl

Materiały źródłowe

  • #1 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The World Health Organisation estimates that by 2030 there will be approximately 350 million people with type 2 diabetes. […] Early identification of patients with undiagnosed type 2 diabetes or those at an increased risk of developing type 2 diabetes is an important challenge. […] We sought to systematically review and critically assess the conduct and reporting of methods used to develop risk prediction models for predicting the risk of having undiagnosed (prevalent) or future risk of developing (incident) type 2 diabetes in adults. […] Risk prediction models have considerable potential to contribute to the decision-making process regarding the clinical management of a patient. […] Healthcare interventions or lifestyle changes can then be targeted towards those at an increased risk of developing a disease.
  • #2 Diabetes Facts and Figures | International Diabetes Federation
    https://idf.org/about-diabetes/diabetes-facts-figures/
    Over 90% of people with diabetes have type 2 diabetes, which is driven by socio-economic, demographic, environmental, and genetic factors. […] However, it is possible to reduce the impact of diabetes by taking preventive measures for type 2 diabetes and providing early diagnosis and proper care for all types of diabetes. These measures can help people living with the condition avoid or delay complications.
  • #3 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The World Health Organisation estimates that by 2030 there will be approximately 350 million people with type 2 diabetes. […] Early identification of patients with undiagnosed type 2 diabetes or those at an increased risk of developing type 2 diabetes is an important challenge. […] We sought to systematically review and critically assess the conduct and reporting of methods used to develop risk prediction models for predicting the risk of having undiagnosed (prevalent) or future risk of developing (incident) type 2 diabetes in adults. […] Risk prediction models have considerable potential to contribute to the decision-making process regarding the clinical management of a patient. […] Healthcare interventions or lifestyle changes can then be targeted towards those at an increased risk of developing a disease.
  • #4 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The World Health Organisation estimates that by 2030 there will be approximately 350 million people with type 2 diabetes. […] Early identification of patients with undiagnosed type 2 diabetes or those at an increased risk of developing type 2 diabetes is an important challenge. […] We sought to systematically review and critically assess the conduct and reporting of methods used to develop risk prediction models for predicting the risk of having undiagnosed (prevalent) or future risk of developing (incident) type 2 diabetes in adults. […] Risk prediction models have considerable potential to contribute to the decision-making process regarding the clinical management of a patient. […] Healthcare interventions or lifestyle changes can then be targeted towards those at an increased risk of developing a disease.
  • #5 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark | BMJ Open Diabetes Research & Care
    https://drc.bmj.com/content/10/5/e002946
    The use of prognostic prediction models can aid in identifying individuals who will develop type 2 diabetes, allowing preventive interventions to be targeted more effectively. […] We showed that one in five individuals from our population will progress to HbA1c-defined type 2 diabetes within 5 years after their first HbA1c-defined pre-diabetes diagnosis, and that one in nine will initiate glucose-lowering treatment within the same period. […] In addition to age, sex, metabolic factors and pre-existing comorbidities, we found that self-rated health, lifestyle, and existence of a social network are important predictors of the progression to type 2 diabetes. […] Although we could identify individuals with pre-diabetes who were at high risk, the AUCts were modest at only 73 (95% CI 71 to 74) for HbA1c-defined type 2 diabetes and 79 (95% CI 78 to 81) for glucose-lowering treatment initiation.
  • #6 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ
    https://www.bmj.com/content/345/bmj.e5900
    Objective To identify existing prediction models for the risk of development of type 2 diabetes and to externally validate them in a large independent cohort. […] Most basic prediction models can identify people at high risk of developing diabetes in a time frame of five to 10 years. Models including biomarkers classified cases slightly better than basic ones. Most models overestimated the actual risk of diabetes. Existing prediction models therefore perform well to identify those at high risk, but cannot sufficiently quantify actual risk of future diabetes. […] An evaluation of the performance of 25 prediction models for type 2 diabetes in an independent Dutch cohort with over 10 years of follow-up showed that basic models perform similarly well in identifying individuals at high and low risk of developing diabetes. The performance was slightly better for extended models that included conventional biomarkers.
  • #7 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ
    https://www.bmj.com/content/345/bmj.e5900
    All except two prediction models overestimated the absolute risk of diabetes in our validation dataset, which can partly be explained by the difference in incidence of diabetes between development and validation populations. To account for this, we adjusted the models for difference in incidence, resulting in much better calibration. Significant deviations between the predicted and observed risks, however, remained for most models. […] Results from our study show that prediction models perform well to identify those at high risk of future diabetes, being a first prerequisite for use of such models in practice as currently recommended.
  • #8 Prediction models for risk of developing type 2 diabetes: systematic literature search and independent external validation study | The BMJ
    https://www.bmj.com/content/345/bmj.e5900
    All except two prediction models overestimated the absolute risk of diabetes in our validation dataset, which can partly be explained by the difference in incidence of diabetes between development and validation populations. To account for this, we adjusted the models for difference in incidence, resulting in much better calibration. Significant deviations between the predicted and observed risks, however, remained for most models. […] Results from our study show that prediction models perform well to identify those at high risk of future diabetes, being a first prerequisite for use of such models in practice as currently recommended.
  • #9 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark
    https://pmc.ncbi.nlm.nih.gov/articles/PMC9486231/
    Pre-diabetes increases the risk of type 2 diabetes, but data are sparse on predictors in a population-based clinical setting. We aimed to develop and validate prediction models for 5-year risks of progressing to type 2 diabetes among individuals with incident HbA1c-defined pre-diabetes. […] The final prediction model included HbA1c, age, sex, body mass index (BMI), any antihypertensive drug use, pancreatic disease, cancer, self-reported diet, doctors advice to lose weight or change dietary habits, having someone to talk to, and self-rated health. […] In addition to well-known pre-diabetes predictors such as age, sex, and BMI, we found that measures of self-rated lifestyle, health, and social support are important and modifiable predictors for diabetes. […] One in five individuals with pre-diabetes will progress to HbA1c-defined diabetes within 5 years.
  • #10 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark | BMJ Open Diabetes Research & Care
    https://drc.bmj.com/content/10/5/e002946
    The use of prognostic prediction models can aid in identifying individuals who will develop type 2 diabetes, allowing preventive interventions to be targeted more effectively. […] We showed that one in five individuals from our population will progress to HbA1c-defined type 2 diabetes within 5 years after their first HbA1c-defined pre-diabetes diagnosis, and that one in nine will initiate glucose-lowering treatment within the same period. […] In addition to age, sex, metabolic factors and pre-existing comorbidities, we found that self-rated health, lifestyle, and existence of a social network are important predictors of the progression to type 2 diabetes. […] Although we could identify individuals with pre-diabetes who were at high risk, the AUCts were modest at only 73 (95% CI 71 to 74) for HbA1c-defined type 2 diabetes and 79 (95% CI 78 to 81) for glucose-lowering treatment initiation.
  • #11 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark
    https://pmc.ncbi.nlm.nih.gov/articles/PMC9486231/
    Pre-diabetes increases the risk of type 2 diabetes, but data are sparse on predictors in a population-based clinical setting. We aimed to develop and validate prediction models for 5-year risks of progressing to type 2 diabetes among individuals with incident HbA1c-defined pre-diabetes. […] The final prediction model included HbA1c, age, sex, body mass index (BMI), any antihypertensive drug use, pancreatic disease, cancer, self-reported diet, doctors advice to lose weight or change dietary habits, having someone to talk to, and self-rated health. […] In addition to well-known pre-diabetes predictors such as age, sex, and BMI, we found that measures of self-rated lifestyle, health, and social support are important and modifiable predictors for diabetes. […] One in five individuals with pre-diabetes will progress to HbA1c-defined diabetes within 5 years.
  • #12 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    Type II diabetes mellitus (T2DM) is a rising global health burden due to its rapidly increasing prevalence worldwide, and can result in serious complications. Therefore, it is of utmost importance to identify individuals at risk as early as possible to avoid long-term T2DM complications. […] The feature importance scores identified a novel combination of biomarkers, including interleukin-10 (IL-10), 8-isoprostane, humanin (HN), and oxidized glutathione (GSSG), which were revealed to be more influential than the traditional biomarkers in the outcome prediction. […] Early identification of individuals at risk of developing T2DM is a priority for the prevention of long-term disease complications. […] The inclusion of biomarkers of OS, inflammation and MD improved the performance across all metrics in comparison to predictive modelling with only traditional biomarkers of BGL, BMI and triglycerides. The greatest boost in performance was observed for recall and F1-scores.
  • #13 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    The top five predictors in terms of DIFFI scores were IL-10, 8-isoprostane, GSSG, HN and P66Shc, while the lowest scores were obtained by BGL and triglycerides, further highlighting the potential role of these novel biomarkers for ML prediction of T2DM development. […] Based on the results of this study, various conclusions can be inferred. First, typical monitoring of T2DM risk through BGL may not provide a comprehensive picture of T2DM disease progression. Influential biomarkers identified were IL-10, 8-isoprostane, GSSG, HN and P66Shc, revealing the potential for biomarkers of inflammation, OD and MD to serve as a guide for targeted, personalized intervention in the prevention of T2DM incidence.
  • #14
    https://link.springer.com/article/10.1007/s13755-021-00168-2
    Type 2 Diabetes (T2D) is a chronic disease characterized by abnormally high blood glucose levels due to insulin resistance and reduced pancreatic insulin production. The challenge of this work is to identify T2D-associated features that can distinguish T2D sub-types for prognosis and treatment purposes. […] Of the features identified by these machine learning models, glucose levels, body mass index, diabetes pedigree function, and age were consistently identified as the best and most frequently accurate outcome predictors. These results indicate the utility of ML methods in constructing improved prediction models for T2D and successfully identified outcome predictors for this Pima Indian population. […] Given the wide variety of presentation and development of comorbidities in T2D, treatment and care of patients can be greatly improved if the prognostic signs are used to better sub-categorize T2D patients.
  • #15 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark
    https://pmc.ncbi.nlm.nih.gov/articles/PMC9486231/
    Pre-diabetes increases the risk of type 2 diabetes, but data are sparse on predictors in a population-based clinical setting. We aimed to develop and validate prediction models for 5-year risks of progressing to type 2 diabetes among individuals with incident HbA1c-defined pre-diabetes. […] The final prediction model included HbA1c, age, sex, body mass index (BMI), any antihypertensive drug use, pancreatic disease, cancer, self-reported diet, doctors advice to lose weight or change dietary habits, having someone to talk to, and self-rated health. […] In addition to well-known pre-diabetes predictors such as age, sex, and BMI, we found that measures of self-rated lifestyle, health, and social support are important and modifiable predictors for diabetes. […] One in five individuals with pre-diabetes will progress to HbA1c-defined diabetes within 5 years.
  • #16 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark | BMJ Open Diabetes Research & Care
    https://drc.bmj.com/content/10/5/e002946
    Pre-diabetes increases the risk of type 2 diabetes, but data are sparse on predictors in a population-based clinical setting. […] We aimed to develop and validate prediction models for 5-year risks of progressing to type 2 diabetes among individuals with incident HbA1c-defined pre-diabetes. […] During a median follow-up of 2.7 years, 11.8% progressed to type 2 diabetes and 10.1% died. […] In addition to well-known pre-diabetes predictors such as age, sex, and BMI, we found that measures of self-rated lifestyle, health, and social support are important and modifiable predictors for diabetes. […] Our model had an acceptable discriminative ability and was well calibrated. […] One in five individuals with pre-diabetes will progress to HbA1c-defined diabetes within 5 years. […] Although we identified individuals with pre-diabetes who were at high risk, the time-dependent area under the curve was only 73 (95% CI 71 to 74) for HbA1c-defined diabetes.
  • #17 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark | BMJ Open Diabetes Research & Care
    https://drc.bmj.com/content/10/5/e002946
    Pre-diabetes increases the risk of type 2 diabetes, but data are sparse on predictors in a population-based clinical setting. […] We aimed to develop and validate prediction models for 5-year risks of progressing to type 2 diabetes among individuals with incident HbA1c-defined pre-diabetes. […] During a median follow-up of 2.7 years, 11.8% progressed to type 2 diabetes and 10.1% died. […] In addition to well-known pre-diabetes predictors such as age, sex, and BMI, we found that measures of self-rated lifestyle, health, and social support are important and modifiable predictors for diabetes. […] Our model had an acceptable discriminative ability and was well calibrated. […] One in five individuals with pre-diabetes will progress to HbA1c-defined diabetes within 5 years. […] Although we identified individuals with pre-diabetes who were at high risk, the time-dependent area under the curve was only 73 (95% CI 71 to 74) for HbA1c-defined diabetes.
  • #18 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark
    https://pmc.ncbi.nlm.nih.gov/articles/PMC9486231/
    Although we identified individuals with pre-diabetes who were at high risk, the time-dependent area under the curve was only 73 (95% CI 71 to 74) for HbA1c-defined diabetes. […] The use of prognostic prediction models can aid in identifying individuals who will develop type 2 diabetes, allowing preventive interventions to be targeted more effectively. […] We showed that one in five individuals from our population will progress to HbA1c-defined type 2 diabetes within 5 years after their first HbA1c-defined pre-diabetes diagnosis, and that one in nine will initiate glucose-lowering treatment within the same period. […] Although we could identify individuals with pre-diabetes who were at high risk, the AUCts were modest at only 73 (95% CI 71 to 74) for HbA1c-defined type 2 diabetes and 79 (95% CI 78 to 81) for glucose-lowering treatment initiation.
  • #19 Development of a 5-year risk prediction model for type 2 diabetes in individuals with incident HbA1c-defined pre-diabetes in Denmark | BMJ Open Diabetes Research & Care
    https://drc.bmj.com/content/10/5/e002946
    The use of prognostic prediction models can aid in identifying individuals who will develop type 2 diabetes, allowing preventive interventions to be targeted more effectively. […] We showed that one in five individuals from our population will progress to HbA1c-defined type 2 diabetes within 5 years after their first HbA1c-defined pre-diabetes diagnosis, and that one in nine will initiate glucose-lowering treatment within the same period. […] In addition to age, sex, metabolic factors and pre-existing comorbidities, we found that self-rated health, lifestyle, and existence of a social network are important predictors of the progression to type 2 diabetes. […] Although we could identify individuals with pre-diabetes who were at high risk, the AUCts were modest at only 73 (95% CI 71 to 74) for HbA1c-defined type 2 diabetes and 79 (95% CI 78 to 81) for glucose-lowering treatment initiation.
  • #20 Type 2 Diabetes: What It Is, Causes, Symptoms & Treatment
    https://my.clevelandclinic.org/health/diseases/21501-type-2-diabetes
    If you have Type 2 diabetes, your outlook depends on several factors, like: […] Untreated or undermanaged T2D can lead to a range of health conditions. […] Potential complications of Type 2 diabetes include: […] Hyperosmolar hyperglycemic state (HHS) is a life-threatening complication of Type 2 diabetes. […] Type 2 diabetes is a chronic (long-term) disease, which means you must manage it for the rest of your life.
  • #21 Type 2 Diabetes: What It Is, Causes, Symptoms & Treatment
    https://my.clevelandclinic.org/health/diseases/21501-type-2-diabetes
    If you have Type 2 diabetes, your outlook depends on several factors, like: […] Untreated or undermanaged T2D can lead to a range of health conditions. […] Potential complications of Type 2 diabetes include: […] Hyperosmolar hyperglycemic state (HHS) is a life-threatening complication of Type 2 diabetes. […] Type 2 diabetes is a chronic (long-term) disease, which means you must manage it for the rest of your life.
  • #22 Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care setting | medRxiv
    https://www.medrxiv.org/content/10.1101/2020.10.08.20209015v3.full-text
    Objective To compare performance of general and diabetes specific cardiovascular risk prediction scores in type 2 diabetes patients (T2DM). […] Results We identified 22 scores: 11 derived in the general population, 9 in only T2DM patients, and 2 that excluded T2DM patients. Over 10 years follow-up, 63,000 events occurred. The RECODE score, derived in people with T2DM, performed best for both CVD (c-statistic 0.731 (0.728,0.734), and CVD+ (0.732 (0.729,0.735)). Overall, neither derivation population, nor original predicted outcome influenced performance. Calibration slopes (1 indicates perfect calibration) ranged from 0.38 (95%CI 0.37;0.39) to 1.05 (95%CI 1.03;1.07). A simple, population specific recalibration process considerably improved performance, ranging between 0.98 and 1.03. Risk scores performed badly in people with pre-existing CVD (c-statistic 0.55). Scores with more predictors did not perform better: for CVD+ QRISK3 (19 variables) c-statistic 0.69 (95%CI 0.68;0.69), compared to CHD Basic (8 variables) 0.71 (95%CI 0.70; 0.71).
  • #23 Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care setting | medRxiv
    https://www.medrxiv.org/content/10.1101/2020.10.08.20209015v3.full-text
    Objective To compare performance of general and diabetes specific cardiovascular risk prediction scores in type 2 diabetes patients (T2DM). […] Results We identified 22 scores: 11 derived in the general population, 9 in only T2DM patients, and 2 that excluded T2DM patients. Over 10 years follow-up, 63,000 events occurred. The RECODE score, derived in people with T2DM, performed best for both CVD (c-statistic 0.731 (0.728,0.734), and CVD+ (0.732 (0.729,0.735)). Overall, neither derivation population, nor original predicted outcome influenced performance. Calibration slopes (1 indicates perfect calibration) ranged from 0.38 (95%CI 0.37;0.39) to 1.05 (95%CI 1.03;1.07). A simple, population specific recalibration process considerably improved performance, ranging between 0.98 and 1.03. Risk scores performed badly in people with pre-existing CVD (c-statistic 0.55). Scores with more predictors did not perform better: for CVD+ QRISK3 (19 variables) c-statistic 0.69 (95%CI 0.68;0.69), compared to CHD Basic (8 variables) 0.71 (95%CI 0.70; 0.71).
  • #24 Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care setting | medRxiv
    https://www.medrxiv.org/content/10.1101/2020.10.08.20209015v3.full-text
    Objective To compare performance of general and diabetes specific cardiovascular risk prediction scores in type 2 diabetes patients (T2DM). […] Results We identified 22 scores: 11 derived in the general population, 9 in only T2DM patients, and 2 that excluded T2DM patients. Over 10 years follow-up, 63,000 events occurred. The RECODE score, derived in people with T2DM, performed best for both CVD (c-statistic 0.731 (0.728,0.734), and CVD+ (0.732 (0.729,0.735)). Overall, neither derivation population, nor original predicted outcome influenced performance. Calibration slopes (1 indicates perfect calibration) ranged from 0.38 (95%CI 0.37;0.39) to 1.05 (95%CI 1.03;1.07). A simple, population specific recalibration process considerably improved performance, ranging between 0.98 and 1.03. Risk scores performed badly in people with pre-existing CVD (c-statistic 0.55). Scores with more predictors did not perform better: for CVD+ QRISK3 (19 variables) c-statistic 0.69 (95%CI 0.68;0.69), compared to CHD Basic (8 variables) 0.71 (95%CI 0.70; 0.71).
  • #25 Cardiovascular risk prediction in type 2 diabetes: a comparison of 22 risk scores in primary care setting | medRxiv
    https://www.medrxiv.org/content/10.1101/2020.10.08.20209015v3.full-text
    Objective To compare performance of general and diabetes specific cardiovascular risk prediction scores in type 2 diabetes patients (T2DM). […] Results We identified 22 scores: 11 derived in the general population, 9 in only T2DM patients, and 2 that excluded T2DM patients. Over 10 years follow-up, 63,000 events occurred. The RECODE score, derived in people with T2DM, performed best for both CVD (c-statistic 0.731 (0.728,0.734), and CVD+ (0.732 (0.729,0.735)). Overall, neither derivation population, nor original predicted outcome influenced performance. Calibration slopes (1 indicates perfect calibration) ranged from 0.38 (95%CI 0.37;0.39) to 1.05 (95%CI 1.03;1.07). A simple, population specific recalibration process considerably improved performance, ranging between 0.98 and 1.03. Risk scores performed badly in people with pre-existing CVD (c-statistic 0.55). Scores with more predictors did not perform better: for CVD+ QRISK3 (19 variables) c-statistic 0.69 (95%CI 0.68;0.69), compared to CHD Basic (8 variables) 0.71 (95%CI 0.70; 0.71).
  • #26 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    Treatment quality indicators measuring lipid- and albuminuria-lowering treatment status are valid quality measures, since they predict a lower risk of cardiovascular events and mortality in patients with diabetes. […] The quality indicators for glucose-lowering treatment should only be used for restricted populations with elevated HbA1c levels. […] Intriguingly, the tested indicators for blood pressure-lowering treatment did not predict patient outcomes. […] This study shows that the quality indicators measuring current treatment status with lipid- and albuminuria-lowering drugs predicted a lower risk of hard cardiovascular outcomes in patients with diabetes in general practice. […] For the indicators measuring treatment intensification, only the one focusing on glucose lowering treatment intensification predicted a lower risk of hard outcomes.
  • #27 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    Treatment quality indicators measuring lipid- and albuminuria-lowering treatment status are valid quality measures, since they predict a lower risk of cardiovascular events and mortality in patients with diabetes. […] The quality indicators for glucose-lowering treatment should only be used for restricted populations with elevated HbA1c levels. […] Intriguingly, the tested indicators for blood pressure-lowering treatment did not predict patient outcomes. […] This study shows that the quality indicators measuring current treatment status with lipid- and albuminuria-lowering drugs predicted a lower risk of hard cardiovascular outcomes in patients with diabetes in general practice. […] For the indicators measuring treatment intensification, only the one focusing on glucose lowering treatment intensification predicted a lower risk of hard outcomes.
  • #28 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    Treatment quality indicators measuring lipid- and albuminuria-lowering treatment status are valid quality measures, since they predict a lower risk of cardiovascular events and mortality in patients with diabetes. […] The quality indicators for glucose-lowering treatment should only be used for restricted populations with elevated HbA1c levels. […] Intriguingly, the tested indicators for blood pressure-lowering treatment did not predict patient outcomes. […] This study shows that the quality indicators measuring current treatment status with lipid- and albuminuria-lowering drugs predicted a lower risk of hard cardiovascular outcomes in patients with diabetes in general practice. […] For the indicators measuring treatment intensification, only the one focusing on glucose lowering treatment intensification predicted a lower risk of hard outcomes.
  • #29 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    None of the quality indicators measuring blood pressure lowering treatment or treatment intensification were predictive of hard outcomes. […] The quality indicators measuring lipid- and albuminuria-lowering treatment status can be considered for implementation into quality indicator sets, since these indicators appear to result in less cardiovascular outcomes. […] The indicator measuring glucose lowering treatment status should be restricted to include only patients with an elevated HbA1c. […] The indicators measuring blood pressure lowering treatment status cannot be used as such, since they are not related to cardiovascular outcomes. […] To measure quality of blood pressure lowering treatment, the use of indicators assessing treatment over time needs further exploration.
  • #30 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    None of the quality indicators measuring blood pressure lowering treatment or treatment intensification were predictive of hard outcomes. […] The quality indicators measuring lipid- and albuminuria-lowering treatment status can be considered for implementation into quality indicator sets, since these indicators appear to result in less cardiovascular outcomes. […] The indicator measuring glucose lowering treatment status should be restricted to include only patients with an elevated HbA1c. […] The indicators measuring blood pressure lowering treatment status cannot be used as such, since they are not related to cardiovascular outcomes. […] To measure quality of blood pressure lowering treatment, the use of indicators assessing treatment over time needs further exploration.
  • #31 Do Treatment Quality Indicators Predict Cardiovascular Outcomes in Patients with Diabetes? | PLOS One
    https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0078821
    None of the quality indicators measuring blood pressure lowering treatment or treatment intensification were predictive of hard outcomes. […] The quality indicators measuring lipid- and albuminuria-lowering treatment status can be considered for implementation into quality indicator sets, since these indicators appear to result in less cardiovascular outcomes. […] The indicator measuring glucose lowering treatment status should be restricted to include only patients with an elevated HbA1c. […] The indicators measuring blood pressure lowering treatment status cannot be used as such, since they are not related to cardiovascular outcomes. […] To measure quality of blood pressure lowering treatment, the use of indicators assessing treatment over time needs further exploration.
  • #32 [2301.03093] Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers
    https://arxiv.org/abs/2301.03093
    Type 2 Diabetes is a fast-growing, chronic metabolic disorder due to imbalanced insulin […] this research is a comparative study of seven machine learning classifiers and an artificial neural network method to prognosticate the detection and treatment of diabetes with high accuracy […] deep ANN which outperforms with 95.14% accuracy among all other tested machine learning […] hope our high-performing model can be used by hospitals to predict diabetes and drive research into more accurate prediction models.
  • #33 Prognosis and Treatment Prediction of Type-2 Diabetes Using Deep Neural Network and Machine Learning Classifiers – ADS
    https://ui.adsabs.harvard.edu/abs/2023arXiv230103093K/abstract
    Type 2 Diabetes is a fast-growing, chronic metabolic disorder due to imbalanced insulin activity. The motion of this research is a comparative study of seven machine learning classifiers and an artificial neural network method to prognosticate the detection and treatment of diabetes with high accuracy, in order to identify and treat diabetes patients at an early age. We use performance measures such as accuracy and precision to find out the best algorithm deep ANN which outperforms with 95.14% accuracy among all other tested machine learning classifiers. We hope our high-performing model can be used by hospitals to predict diabetes and drive research into more accurate prediction models.
  • #34 Early detection of type 2 diabetes mellitus using machine learning-based prediction models | Scientific Reports
    https://www.nature.com/articles/s41598-020-68771-z
    Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. […] Our results show no clinically relevant improvement when more sophisticated prediction models were used. […] The aim of this study was to investigate whether novel machine learning-based approaches offered any advantages over standard regression techniques in early prediction of impaired fasting glucose (IFG) and fasting plasma glucose level (FPGL) values. […] Our results found no clinically relevant improvement when employing machine learning-based models over the more conventional regression models in terms of predictive performance. […] The opportunity of updating models arises as additional routine data become available over time.
  • #35 Early detection of type 2 diabetes mellitus using machine learning-based prediction models | Scientific Reports
    https://www.nature.com/articles/s41598-020-68771-z
    Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. […] Our results show no clinically relevant improvement when more sophisticated prediction models were used. […] The aim of this study was to investigate whether novel machine learning-based approaches offered any advantages over standard regression techniques in early prediction of impaired fasting glucose (IFG) and fasting plasma glucose level (FPGL) values. […] Our results found no clinically relevant improvement when employing machine learning-based models over the more conventional regression models in terms of predictive performance. […] The opportunity of updating models arises as additional routine data become available over time.
  • #36 Early detection of type 2 diabetes mellitus using machine learning-based prediction models | Scientific Reports
    https://www.nature.com/articles/s41598-020-68771-z
    Most screening tests for T2DM in use today were developed using multivariate regression methods that are often further simplified to allow transformation into a scoring formula. […] Our results show no clinically relevant improvement when more sophisticated prediction models were used. […] The aim of this study was to investigate whether novel machine learning-based approaches offered any advantages over standard regression techniques in early prediction of impaired fasting glucose (IFG) and fasting plasma glucose level (FPGL) values. […] Our results found no clinically relevant improvement when employing machine learning-based models over the more conventional regression models in terms of predictive performance. […] The opportunity of updating models arises as additional routine data become available over time.
  • #37
    https://link.springer.com/article/10.1007/s13755-021-00168-2
    The classification results of this work are represented with the resampling distribution of summary statistics more accurately. This combination can identify the top performing machine learning model from a range of different viewpoints. […] In this study, we analyzed PIDD and its sub-datasets using various classifiers to identify the best classifier based on experiment results. […] Among all classifiers, GAMBoost and GAMLOESS provided the best outcomes in this analysis. That is to say that, GAMBoost gave a better performance than GAMLOESS for accuracy, sensitivity while, GAMLOESS showed better results for AUROC and specificity. […] In this work, we investigated the PIDD T2D dataset using various statistical, machine learning and visualization techniques to determine the ranking of classifiers and feature subsets. We found that GAMLOESS was the top ranked classifier and FS5 was the most significant feature subset for achieving the best classifications and analyzing this disease.
  • #38 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. […] We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. […] Despite the large number of risk prediction models being developed, only a very small minority end up being routinely used in clinical practice. […] There is a growing concern that the majority of risk prediction models are poorly developed because they are based on a small and inappropriate selection of the cohort, questionable handling of continuous risk predictors, inappropriate treatment of missing data, use of flawed or unsuitable statistical methods and, ultimately, a lack of transparent reporting of the steps taken to derive the model. […] This systematic review of 39 published studies highlights numerous methodological deficiencies and a generally poor level of reporting in studies in which risk prediction models were developed for the detection of prevalent or incident type 2 diabetes.
  • #39 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. […] We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. […] Despite the large number of risk prediction models being developed, only a very small minority end up being routinely used in clinical practice. […] There is a growing concern that the majority of risk prediction models are poorly developed because they are based on a small and inappropriate selection of the cohort, questionable handling of continuous risk predictors, inappropriate treatment of missing data, use of flawed or unsuitable statistical methods and, ultimately, a lack of transparent reporting of the steps taken to derive the model. […] This systematic review of 39 published studies highlights numerous methodological deficiencies and a generally poor level of reporting in studies in which risk prediction models were developed for the detection of prevalent or incident type 2 diabetes.
  • #40 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. […] We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. […] Despite the large number of risk prediction models being developed, only a very small minority end up being routinely used in clinical practice. […] There is a growing concern that the majority of risk prediction models are poorly developed because they are based on a small and inappropriate selection of the cohort, questionable handling of continuous risk predictors, inappropriate treatment of missing data, use of flawed or unsuitable statistical methods and, ultimately, a lack of transparent reporting of the steps taken to derive the model. […] This systematic review of 39 published studies highlights numerous methodological deficiencies and a generally poor level of reporting in studies in which risk prediction models were developed for the detection of prevalent or incident type 2 diabetes.
  • #41 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. […] We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. […] Despite the large number of risk prediction models being developed, only a very small minority end up being routinely used in clinical practice. […] There is a growing concern that the majority of risk prediction models are poorly developed because they are based on a small and inappropriate selection of the cohort, questionable handling of continuous risk predictors, inappropriate treatment of missing data, use of flawed or unsuitable statistical methods and, ultimately, a lack of transparent reporting of the steps taken to derive the model. […] This systematic review of 39 published studies highlights numerous methodological deficiencies and a generally poor level of reporting in studies in which risk prediction models were developed for the detection of prevalent or incident type 2 diabetes.
  • #42 Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting
    https://pmc.ncbi.nlm.nih.gov/articles/PMC3180398/
    The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. […] We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. […] Despite the large number of risk prediction models being developed, only a very small minority end up being routinely used in clinical practice. […] There is a growing concern that the majority of risk prediction models are poorly developed because they are based on a small and inappropriate selection of the cohort, questionable handling of continuous risk predictors, inappropriate treatment of missing data, use of flawed or unsuitable statistical methods and, ultimately, a lack of transparent reporting of the steps taken to derive the model. […] This systematic review of 39 published studies highlights numerous methodological deficiencies and a generally poor level of reporting in studies in which risk prediction models were developed for the detection of prevalent or incident type 2 diabetes.
  • #43 Prediction Models for Type 2 Diabetes Risk in the General Population: A Systematic Review of Observational Studies
    https://brieflands.com/articles/ijem-109206
    Generally, despite its new risk factors or new methodological aspects, the newly developed model did not increase our capability in screening/predicting T2DM, mainly in the analysis part. It was due to the lack of external validation of the prediction models. […] The overall judgment of ROB assessment is shown in Figure 3. Low ROB was noted in three domains of participants, predictors, and outcomes for both I-T2DM and U-T2DM. […] The strength of this study is that it was reported in accordance with the PRISMA-ScR checklist. This review also included a comprehensive report of model development (e.g., the outcome definition, variable selection, statistical analysis, and treatment of continuous variables) and validation (e.g., calibration and net benefit) requirements according to the TRIPOD guideline. […] It poses the question whether we could rely on the current prediction models or we should develop new models. Another major concern is that a newly developed model can be easily disregarded if it has no added value for health policymakers or clinicians.
  • #44 Prediction Models for Type 2 Diabetes Risk in the General Population: A Systematic Review of Observational Studies
    https://brieflands.com/articles/ijem-109206
    Generally, despite its new risk factors or new methodological aspects, the newly developed model did not increase our capability in screening/predicting T2DM, mainly in the analysis part. It was due to the lack of external validation of the prediction models. […] The overall judgment of ROB assessment is shown in Figure 3. Low ROB was noted in three domains of participants, predictors, and outcomes for both I-T2DM and U-T2DM. […] The strength of this study is that it was reported in accordance with the PRISMA-ScR checklist. This review also included a comprehensive report of model development (e.g., the outcome definition, variable selection, statistical analysis, and treatment of continuous variables) and validation (e.g., calibration and net benefit) requirements according to the TRIPOD guideline. […] It poses the question whether we could rely on the current prediction models or we should develop new models. Another major concern is that a newly developed model can be easily disregarded if it has no added value for health policymakers or clinicians.
  • #45
    https://link.springer.com/article/10.1007/s13755-021-00168-2
    Type 2 Diabetes (T2D) is a chronic disease characterized by abnormally high blood glucose levels due to insulin resistance and reduced pancreatic insulin production. The challenge of this work is to identify T2D-associated features that can distinguish T2D sub-types for prognosis and treatment purposes. […] Of the features identified by these machine learning models, glucose levels, body mass index, diabetes pedigree function, and age were consistently identified as the best and most frequently accurate outcome predictors. These results indicate the utility of ML methods in constructing improved prediction models for T2D and successfully identified outcome predictors for this Pima Indian population. […] Given the wide variety of presentation and development of comorbidities in T2D, treatment and care of patients can be greatly improved if the prognostic signs are used to better sub-categorize T2D patients.
  • #46 Diabetes Facts and Figures | International Diabetes Federation
    https://idf.org/about-diabetes/diabetes-facts-figures/
    Over 90% of people with diabetes have type 2 diabetes, which is driven by socio-economic, demographic, environmental, and genetic factors. […] However, it is possible to reduce the impact of diabetes by taking preventive measures for type 2 diabetes and providing early diagnosis and proper care for all types of diabetes. These measures can help people living with the condition avoid or delay complications.
  • #47 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    Type II diabetes mellitus (T2DM) is a rising global health burden due to its rapidly increasing prevalence worldwide, and can result in serious complications. Therefore, it is of utmost importance to identify individuals at risk as early as possible to avoid long-term T2DM complications. […] The feature importance scores identified a novel combination of biomarkers, including interleukin-10 (IL-10), 8-isoprostane, humanin (HN), and oxidized glutathione (GSSG), which were revealed to be more influential than the traditional biomarkers in the outcome prediction. […] Early identification of individuals at risk of developing T2DM is a priority for the prevention of long-term disease complications. […] The inclusion of biomarkers of OS, inflammation and MD improved the performance across all metrics in comparison to predictive modelling with only traditional biomarkers of BGL, BMI and triglycerides. The greatest boost in performance was observed for recall and F1-scores.
  • #48 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    Type II diabetes mellitus (T2DM) is a rising global health burden due to its rapidly increasing prevalence worldwide, and can result in serious complications. Therefore, it is of utmost importance to identify individuals at risk as early as possible to avoid long-term T2DM complications. […] The feature importance scores identified a novel combination of biomarkers, including interleukin-10 (IL-10), 8-isoprostane, humanin (HN), and oxidized glutathione (GSSG), which were revealed to be more influential than the traditional biomarkers in the outcome prediction. […] Early identification of individuals at risk of developing T2DM is a priority for the prevention of long-term disease complications. […] The inclusion of biomarkers of OS, inflammation and MD improved the performance across all metrics in comparison to predictive modelling with only traditional biomarkers of BGL, BMI and triglycerides. The greatest boost in performance was observed for recall and F1-scores.
  • #49 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    The top five predictors in terms of DIFFI scores were IL-10, 8-isoprostane, GSSG, HN and P66Shc, while the lowest scores were obtained by BGL and triglycerides, further highlighting the potential role of these novel biomarkers for ML prediction of T2DM development. […] Based on the results of this study, various conclusions can be inferred. First, typical monitoring of T2DM risk through BGL may not provide a comprehensive picture of T2DM disease progression. Influential biomarkers identified were IL-10, 8-isoprostane, GSSG, HN and P66Shc, revealing the potential for biomarkers of inflammation, OD and MD to serve as a guide for targeted, personalized intervention in the prevention of T2DM incidence.
  • #50 Exploratory risk prediction of type II diabetes with isolation forests and novel biomarkers | Scientific Reports
    https://www.nature.com/articles/s41598-024-65044-x
    The top five predictors in terms of DIFFI scores were IL-10, 8-isoprostane, GSSG, HN and P66Shc, while the lowest scores were obtained by BGL and triglycerides, further highlighting the potential role of these novel biomarkers for ML prediction of T2DM development. […] Based on the results of this study, various conclusions can be inferred. First, typical monitoring of T2DM risk through BGL may not provide a comprehensive picture of T2DM disease progression. Influential biomarkers identified were IL-10, 8-isoprostane, GSSG, HN and P66Shc, revealing the potential for biomarkers of inflammation, OD and MD to serve as a guide for targeted, personalized intervention in the prevention of T2DM incidence.
  • #51 Establishing a Prediction Model for Weight Loss Outcomes After LSG in | DMSO
    https://www.dovepress.com/establishing-a-prediction-model-for-weight-loss-outcomes-after-lsg-in–peer-reviewed-fulltext-article-DMSO
    In obese Chinese patients with a BMI 32.5 kg/m2, the Inbody-based nomogram integrating REE/BW, FFMI, and WC offers an effective preoperative tool for predicting weight loss outcomes one year after LSG, facilitating surgical planning and postoperative management. […] The five most influential variables were subsequently included in a multivariate logistic regression model, which indicated that REE/BW, FFMI, and WC were independent predictors of weight loss outcomes. […] This study developed a preoperative predictive model for weight loss outcomes following LSG by integrating three key indicators: REE/BW, FFMI, and WC. […] In conclusion, the combination of REE/BW, FFMI, and WC provides a relatively accurate preoperative prediction of weight loss outcomes one year post-LSG.